Update dataset card for 20k subsets

README.md

@@ -845,7 +845,7 @@ dataset_info:
 
 # Language Decoded | Multilingual Code Dataset
 
-> **Note (2026-04-21):** Phase 2 configs have been renamed to `phase-2-the-stack-v1-*` and moved under `data/phase-2-the-stack-v1/`. Existing `load_dataset(...)` calls using the short `condition-*` names will no longer resolve. The short `condition-*` namespace is reserved for Phase 3 builds (streamed from The Stack v2, coming soon). See the _Loading the Dataset_ section below for updated config names.
+> **Note (2026-05-18):** Current Phase 3 configs use the short `condition-*` namespace and include `103k`, `20k`, and `5k` sizes for Conditions 1--2. Phase 2 configs remain available under the `phase-2-the-stack-v1-*` namespace for reproducibility.
 
 Multilingual Python code datasets for the **Language Decoded** project (part of [Cohere's Tiny Aya Expedition](https://aya.for.ai)), investigating whether code's reasoning benefit for language models is **language-dependent** or **structure-dependent**.
 
@@ -870,20 +870,25 @@ Keyword-swapped variants are produced using [Legesher](https://github.com/legesh
 
 ## Available Configs
 
-Each condition is available in two sizes: `-32k` (full filtered corpus, ~31.8k train + ~3.5k validation) and `-5k` (stratified subset, 4.5k train + 500 validation). The `-5k` subsets are used for QLoRA fine-tuning on consumer GPUs.
+Conditions 1--2 are available in three current Phase 3 sizes: `-103k` full corpora, `-20k` random subsets sampled from the corresponding `-103k` config with seed 42, and `-5k` compact subsets. Phase 2 `-32k` configs are still available with the `phase-2-the-stack-v1-*` prefix.
 
-| Config               | Condition   | Language | Description                                                  | Train  | Val   |
-| -------------------- | ----------- | -------- | ------------------------------------------------------------ | ------ | ----- |
-| `condition-1-en-32k` | 1 (control) | English  | Unmodified filtered Python from The Stack Dedup              | 31,818 | 3,536 |
-| `condition-1-en-5k`  | 1 (control) | English  | Stratified 5k subset of condition-1                          | 4,500  | 500   |
-| `condition-2-zh-32k` | 2           | Chinese  | Keyword-swapped Python via Legesher v0.7.3                   | 31,818 | 3,536 |
-| `condition-2-zh-5k`  | 2           | Chinese  | Stratified 5k subset of condition-2-zh                       | 4,500  | 500   |
-| `condition-2-es-32k` | 2           | Spanish  | Keyword-swapped Python via Legesher v0.7.3                   | 31,818 | 3,536 |
-| `condition-2-es-5k`  | 2           | Spanish  | Stratified 5k subset of condition-2-es                       | 4,500  | 500   |
-| `condition-2-ur-32k` | 2           | Urdu     | Keyword-swapped Python via Legesher v0.7.3                   | 31,818 | 3,536 |
-| `condition-2-ur-5k`  | 2           | Urdu     | Stratified 5k subset of condition-2-ur                       | 4,500  | 500   |
-| `condition-3-zh-5k`  | 3           | Chinese  | Blended: 3,486 native Chinese code + 1,514 transpiled Python | 4,500  | 500   |
-| `condition-4-zh-5k`  | 4           | Chinese  | Strictly native Chinese code (no transpiled code)            | 6,553  | 729   |
+| Config                    | Condition   | Language | Description                                      | Train  | Val    |
+| ------------------------- | ----------- | -------- | ------------------------------------------------ | ------ | ------ |
+| `condition-1-en-103k`     | 1 (control) | English  | Unmodified filtered Python                       | 93,549 | 10,395 |
+| `condition-1-en-20k`      | 1 (control) | English  | Random 20k subset of `condition-1-en-103k`       | 18,000 | 2,000  |
+| `condition-1-en-5k`       | 1 (control) | English  | Compact 5k subset                                | 4,500  | 500    |
+| `condition-2-zh-103k`     | 2           | Chinese  | Keyword-swapped Python via Legesher v0.7.3       | 93,547 | 10,395 |
+| `condition-2-zh-20k`      | 2           | Chinese  | Random 20k subset of `condition-2-zh-103k`       | 18,000 | 2,000  |
+| `condition-2-zh-5k`       | 2           | Chinese  | Compact 5k subset                                | 4,500  | 500    |
+| `condition-2-es-103k`     | 2           | Spanish  | Keyword-swapped Python via Legesher v0.7.3       | 93,547 | 10,395 |
+| `condition-2-es-20k`      | 2           | Spanish  | Random 20k subset of `condition-2-es-103k`       | 18,000 | 2,000  |
+| `condition-2-es-5k`       | 2           | Spanish  | Compact 5k subset                                | 4,500  | 500    |
+| `condition-2-ur-103k`     | 2           | Urdu     | Keyword-swapped Python via Legesher v0.7.3       | 93,547 | 10,395 |
+| `condition-2-ur-20k`      | 2           | Urdu     | Random 20k subset of `condition-2-ur-103k`       | 18,000 | 2,000  |
+| `condition-2-ur-5k`       | 2           | Urdu     | Compact 5k subset                                | 4,500  | 500    |
+| `condition-3-zh-5k`       | 3           | Chinese  | Blended: native Chinese code + transpiled Python | 4,500  | 500    |
+| `condition-4-zh-5k`       | 4           | Chinese  | Strictly native Chinese code                     | 6,553  | 729    |
+| `condition-5-ur-5k-c4ai-aya-expanse-32b` | 5 | Urdu | Model-translated Urdu condition                  | 4,024  | 447    |
 
 ## Schema
 
@@ -960,7 +965,13 @@ The Language Decoded experiment uses a ladder of conditions to isolate the mecha
 from datasets import load_dataset
 
 # Load full-size English code (control)
-ds = load_dataset("legesher/language-decoded-data", "condition-1-en-32k")
+ds = load_dataset("legesher/language-decoded-data", "condition-1-en-103k")
+
+# Load random 20k subsets
+ds = load_dataset("legesher/language-decoded-data", "condition-1-en-20k")
+ds = load_dataset("legesher/language-decoded-data", "condition-2-zh-20k")
+ds = load_dataset("legesher/language-decoded-data", "condition-2-es-20k")
+ds = load_dataset("legesher/language-decoded-data", "condition-2-ur-20k")
 
 # Load 5k subset (for QLoRA fine-tuning)
 ds = load_dataset("legesher/language-decoded-data", "condition-1-en-5k")